Diversifying FIDC Portfolios by Risk Source: Looking Beyond the Number of Borrowers

What a large portfolio can conceal

The relevant question is not simply how many receivables the fund holds, but which events could prevent them from becoming cash. Debtor insolvency, a sale that never occurred, an earlier assignment, a commercial discount, collection failure and improper withholding of payments are different loss mechanisms. Some affect individual claims; others can affect many together. Analysis should identify each mechanism, its reach and potential recovery routes.

This perspective changes how portfolio reports should be read. Low concentration by debtor can coexist with heavy dependence on one assignor, collection platform or data source. Conversely, two transactions in the same industry may behave differently because their buyers, contracts and payment arrangements differ. Industry classifications provide a starting point for investigation, rather than a complete diagnosis of shared exposure.

Diversification also cannot replace asset quality. Spreading unsupported claims across many names merely distributes documentary uncertainty. The objective is to combine eligible, sufficiently understood assets while recognizing their connections. This article examines that process for FIDCs and credit managers in Brazil. It does not prescribe personal asset allocation or assume that any particular portfolio suits every investor.

Eligibility comes before combining risks

Start by checking whether the receivables meet the fund's policy, acquisition criteria and applicable limits. CVM Resolution 175, including Annex II, provides a reference framework for Brazilian FIDCs. The structure's documents and the responsibilities of its service providers also matter. A commercially attractive opportunity may be unsuitable for one fund even if another structure could consider it.

An existing receivable differs from expected revenue. A business may have a credible expansion project and forecast future sales, but that forecast is not an established claim against an identified buyer. Even where a contract exists, determine what makes payment due: delivery, acceptance, measurement of completed work, a project milestone or another condition. The portfolio's underlying asset must be clearly identified.

Loans, receivables, equity interests, convertible instruments and donations create different rights and risks. A production loan does not become a trade receivable merely because sales will fund repayment. A donation to a nonprofit is neither an equity investment nor a promise of financial return. A Brazilian individual microentrepreneur registered as a MEI cannot admit equity partners while retaining MEI status. Legal substance must precede the economic narrative.

KEZS connects entrepreneurial projects and potential funders through technology and intermediation. Presentation of a project does not make its claims eligible for a FIDC. Investors conduct their own diligence and make their own decisions; KEZS does not assign a credit score. Organizing information by project may support understanding, but its usefulness depends on evidence quality and the work of those responsible for the transaction.

Build a dependency map, not just a list of entities

A risk map begins with each claim: who owes the money, who originated and assigned it, which activity generates repayment funds, and how payment reaches the fund. Add less visible dependencies, including economic groups, ultimate buyers, delivery regions, essential suppliers and document systems. These relationships may be invisible in a list organized only by CNPJ, Brazil's business registration identifier.

Consider three independent suppliers: one sells food, another packaging and a third maintenance services. Their debtors may operate in different industries but depend on the same retail network for revenue. If that network stops paying, several positions may suffer. Distinguish direct claims against the network from indirect commercial dependence, while ensuring that the same receivable is not counted twice.

Separate credit, legal, operational and liquidity risks without assuming they are independent. A collection system outage may delay reconciliation, obscure overdue balances and consume cash needed for fund obligations. An operational failure can increase credit losses if it prevents timely collection or allows payments to be diverted. The map should show how one kind of failure can spread into another.

A useful map explains both known connections and information gaps. Economic group information may be incomplete, or dependence on a buyer may rest on an unverified declaration. Record confidence levels and the consequences of missing data. Uncertainty about a material connection calls for further investigation or exposure constraints; it is not evidence that the positions are independent.

Existence and ownership require a chain of evidence

Before estimating payment likelihood, establish what is payable and who owns the right. Evidence of existence may require a contract, purchase order, tax invoice, delivery record, measurement of work and acceptance. No single document answers every question. An invoice issued without delivery, for example, does not resolve a dispute over whether the underlying commercial transaction actually occurred.

Ownership requires tracing the chain through which the claim reaches the fund. Who was the original creditor? Who has authority to assign it? Were there intermediate transfers? Does the assignment identify the claims precisely enough? Answers depend on the right and structure involved. Investors need to understand who performs these checks, what evidence supports them and how exceptions are handled.

Evidence becomes stronger when independent sources corroborate one another. An assignor's file can be compared with delivery information and, where appropriate, confirmation obtained from the debtor through a reliable channel. Confirmation still does not eliminate later returns or quality disputes. The evidence must address the particular loss mechanism under review; accumulating documents is not a substitute for establishing relevant facts.

Sampling can support analysis of a large portfolio, but a sample without exceptions does not prove that every claim is valid. Understand how the sample was selected, what it covered, which failures appeared and what followed. A significant exception may justify broader checks, suspension of acquisitions or reassessment of the data source. The response should reflect the potential extent of the problem.

Encumbrances and prior assignments can connect otherwise separate losses

An existing receivable may already have been assigned or may carry encumbrances affecting its availability. Documents, registration records and the assignor's statements may also conflict. Diligence must therefore extend beyond existence and ownership to restrictions, earlier transfers and third-party rights. The relevant procedures depend on the asset, applicable systems and legal analysis of the transaction.

This becomes a concentration risk when many claims share the same origination process. An assignor with weak controls might offer the same receivables to different funders. The resulting problem can affect numerous debtors simultaneously, despite a scattered debtor list. Diversifying buyers does not repair a compromised assignment chain or make unavailable assets available to the fund.

Contractual statements that claims are free of encumbrances are useful, but protection depends on verification and enforceability. An indemnity may recover little if the assignor lacks resources. Repurchase obligations require the same scrutiny: what triggers them, and could the obligated party perform when several irregularities emerge together? A legal promise and financial capacity to honor it are separate considerations.

Consistent controls link individual claim identification, appropriate searches, formalization, record updates and acquisition blocks when conflicts appear. Finding a prior assignment should prompt more than deleting a spreadsheet row. Investigate scope, preserve evidence, assess the existing portfolio and follow the structure's communication and decision procedures. The underlying error may remain active across other claims.

Concentration needs more than one denominator

Debtor concentration measures exposure to that debtor's failure to pay. It does not independently measure reliance on an assignor, economic group, industry, region or collection channel. Every reported dimension needs a stated denominator. A percentage of receivables principal differs from a percentage of total fund assets when the fund also holds cash or other investments.

The measurement point matters too. A portfolio may comply with limits when assets are purchased and become more concentrated as other claims are repaid. If stronger payers exit while troubled claims remain, the residual portfolio changes in quality. Track composition, maturities and arrears over time, and examine the procedures that apply if the portfolio moves outside its limits.

Separate limits do not necessarily capture combined dependencies. A fund may remain below both assignor and industry limits while retaining substantial exposure to one ultimate buyer. Another combination might involve a small assignor, many debtors and collections controlled by that assignor. Managers must assess these intersections within the structure's policy and limits; formal compliance does not replace analysis.

Absolute amounts, percentages and cash consequences give investors a clearer picture. Reporting a group's portfolio share becomes more useful when accompanied by the amount of cash that would fail to arrive if its payments stopped. Avoid double counting: one claim may appear under several risk categories, but it remains one economic exposure when calculating total assets.

A numerical example: portfolio, cash and a shared dependency

Consider a hypothetical fund starting with R$10 million: R$9 million of receivables principal and R$1 million in cash. There are no new acquisitions, contributions or distributions during the exercise. The receivables comprise R$3 million linked to food retail, R$2.5 million to manufacturing, R$2 million to services and R$1.5 million to wholesale trade. Together, these equal exactly R$9 million.

A shared logistics dependency cuts across those groups: R$1.2 million in food retail, R$800,000 in manufacturing and R$600,000 in services. The same event therefore affects R$2.6 million, without adding any assets to the portfolio. This represents approximately 28.9% of receivables principal and 26% of initial assets. The remaining R$6.4 million is assumed unaffected by this specific event.

Assume the contracts generate R$360,000 of financial income if performed in full. To simplify the calculation, every portion of principal repaid carries income equal to 4% of that principal. This is a hypothetical arithmetic assumption, not a market rate or annualized rate. Total expected expenses through final collection are R$60,000. No claim about actual investment performance follows from these assumptions.

With no failures, ending accumulated cash would be R$10.3 million: R$1 million initially held, plus R$9 million of principal, plus R$360,000 of income, less R$60,000 of expenses. Accounting recognition would depend on applicable criteria; this exercise tracks cash receipts and payments. Four industry groups do not yet demonstrate effective diversification, because the common dependency still needs to be assessed.

The payment schedule reveals what the ending balance conceals

Keep the same amounts and divide maturities into three points measured from the start of the exercise. R$3 million of principal falls due on day 30, another R$3 million on day 60 and the final R$3 million on day 90. Each batch carries R$120,000 of contractual income. Expenses are paid in three R$20,000 installments at those same points. Receipts are not reinvested.

On day 30, receipts of R$3.12 million and expenses of R$20,000 increase cash from R$1 million to R$4.1 million. The same movement raises the balance to R$7.2 million on day 60 and R$10.3 million on day 90. These are accumulated cash balances under the stated assumptions, with no additional obligations and no return earned on the cash itself.

Now add a crucial detail: all R$2.6 million exposed to the logistics dependency matures on day 90. The first two batches are entirely outside that event; the third contains R$400,000 unaffected and R$2.6 million affected. Risk source and maturity are concentrated together. Normal collections from the first two batches do not establish that the final batch will behave similarly.

Any fund payment obligation before day 90 would need to be compared with cash available at that point and the structure's rules. A reserve may cover a temporary mismatch, but cannot eliminate a permanent loss. The schedule separates three issues: a claim that does not pay, one that pays late, and a fund obligation due before expected receipts. Each requires a different response.

A consistent adverse scenario, including recovery and costs

Assume the logistics interruption causes commercial disputes and prevents payment of R$2.6 million on day 90. For this stress scenario, assume recovery of 60% of that principal on day 180, with none of its associated contractual income received. The other R$6.4 million pays as scheduled. These assumptions are not forecasts of the frequency, duration or recovery rate of actual events.

The affected R$2.6 million carried R$104,000 of income at the assumed 4%; unaffected claims generate R$256,000. Accumulated cash on day 90 is R$7.596 million: initial cash of R$1 million, plus R$6.4 million principal and R$256,000 income, less R$60,000 expenses. Equivalently, add R$416,000 to the day-60 balance of R$7.2 million and subtract the final R$20,000 expense installment.

On day 180, recovering 60% adds R$1.56 million. Assume another R$40,000 of collection and monitoring expenses is paid by then. Ending cash becomes R$9.116 million. Unrecovered principal on the affected claims is R$1.04 million. After the income actually received and all scenario expenses, the result is a loss of R$884,000 against the initial R$10 million.

The shortfall from the no-failure outcome of R$10.3 million is R$1.184 million: R$1.04 million of unrecovered principal, R$104,000 of missed income and R$40,000 of additional costs. Delay until day 180 also matters, although this exercise assigns no financial value to time. Doing so would require an explicit assumption and care not to double count opportunity cost as an existing cash loss.

Compare protections against the same event

Possible responses to the R$2.6 million concentration include reducing exposure, acquiring claims with different dependencies, obtaining stronger evidence or negotiating contractual protections, subject to the fund's criteria and documents. Match each measure to the risk it addresses. A larger purchase discount may improve the economic margin, but cannot establish that a claim exists or repair defective ownership.

Assess collateral, co-obligations and repurchase promises under the event that affects the debtor. A guarantor dependent on the same retail network may weaken alongside the portfolio. Realizing collateral can take time and incur costs. A well-drafted repurchase obligation may have limited economic value if the assignor faces many demands together. Contractual protection does not necessarily provide immediately available cash.

Within a class, unit subclasses may have different payment priorities and loss-absorption positions under the applicable rules and documents. Explain that order, including the restriction against establishing subordination between different subordinated FIDC subclasses. Separate classes have segregated assets, with no presumed cross-protection. Subordination changes an investor's exposure but does not restore lost portfolio resources. Its effectiveness depends on its size, the class's evolving assets, contractual triggers and compliance conditions, not its name.

Diversification also costs money. More assignors can require additional integrations, checks and collection routines. A small transaction may consume disproportionate effort, while manageable concentration may permit deeper controls. Compare the expected benefit, cost and residual risk of each alternative. The objective is to prevent one failure source from overwhelming loss-bearing capacity, rather than simply maximize the number of positions.

Collections are part of the risk architecture

Collection arrangements determine how a recognized right becomes identifiable cash. Establish who issues payment instructions, receives funds, reconciles receipts and handles exceptions. A debtor may pay correctly while an intermediary withholds the money or allocates it to the wrong claim. That creates operational and potentially financial harm. Diversification analysis must examine the payment route as well as the payer's economic position.

Evaluate a collection account or payment procedure through contracts, movement controls and evidence of actual implementation. A flowchart in a presentation does not prove that the process operates. Useful evidence includes reconciliations, exception logs, records of instruction changes and checks on where funds ultimately arrive. The process must function when records disagree, not merely when every entry matches.

Identify commercial dilution: returns, allowances, offsets, discounts and disputes can reduce collectible amounts without initially appearing as payment defaults. If many claims share a delivery process or an assignor's commercial policy, dilution can rise across them together. Labeling every difference as lateness hides its cause and may delay changes needed at origination. Separate reduced obligations from unpaid amounts that remain due.

Test continuity explicitly. If the collection provider becomes unavailable, can another authorized party reconstruct balances, locate documents and resume contact? This requires permitted data access, clear responsibilities and ways to replace critical routines. Two nominally separate providers offer limited resilience if both rely on the same inaccessible system. Operational diversity must be demonstrated through actual dependencies.

Clear responsibilities reduce blind spots

A structure may involve an administrator, investment manager, custodian where applicable, registration entities, advisers and collection agents. Check their actual duties against relevant rules and contracts rather than assuming every FIDC uses the same arrangement. Investors should understand who authorizes acquisitions, checks the corresponding criteria, retains evidence and responds when a required condition is no longer met.

A responsibility matrix should connect every control to an owner, supporting evidence and an exception procedure. Delivery verification, for example, should produce an identifiable record; an inconsistency should prevent progression as required by established rules; resolution should require the appropriate authority. A general statement that diligence will occur leaves a governance gap if it does not explain process and accountability.

Conflicts arise when a party rewarded for volume also supplies the information used to approve acquisitions. An assignor may minimize returns; an origination provider may favor speed; a collection agent may emphasize signed settlements without reporting cash received. Controls should allow challenges, data review and outcome monitoring that do not depend solely on statements from the party with an interest in the result.

A project-based method may help organize purpose, contracts, participants and documentary progress where information is verifiable and available to authorized users. It does not transfer fund responsibilities to a connection platform. Keep organized documentation, completed verification and investment decisions distinct. They support one another, but completing one stage does not automatically establish that the others were properly completed.

Measure behavior without confusing data with conclusions

A useful dashboard extends beyond outstanding defaults to aging bands, losses, cash recoveries, dilution, concentration, documentary exceptions and reconciliation times. Definitions must remain stable. Lateness measured from original maturity cannot be compared directly with lateness measured from renegotiated maturity without explaining the difference. A methodological change can otherwise create an improvement that exists only in the reported indicator.

Acquisition-vintage analysis can show whether claims bought in one period perform worse than earlier batches. Compare equivalent stages: newly acquired claims have had less time to reveal problems. Portfolio growth can also reduce the percentage overdue without recovering old claims. Absolute amounts, batch histories and write-off criteria help distinguish real improvement from a change in the denominator or reporting population.

History is informative but does not automatically provide probabilities for rare events or structural changes. An assignor with few late payments may never have experienced the logistics failure in the example. Apparent stability may also reflect repeated repurchases of troubled claims. Investigate whether the portfolio performed or problems moved to another party whose financial capacity needs separate assessment.

Technology may improve traceability when records correspond to verifiable events. Blockchain can make certain later alterations harder, but does not prove a sale's truth, legal title, reserve solvency or future payment. Private documents should not be placed on public blockchains. Data protection requires access proportionate to purpose, sharing controls and handling appropriate to the sensitivity of personal and commercial information.

Disagreement and negotiation should lead to verifiable decisions

Suppose the commercial team considers the four industry groups diversified while the risk team highlights R$2.6 million of logistics dependence. The classifications differ; examining the mechanism resolves the disagreement. Ask which deliveries would stop, which debtors could replace the service and which payments are already legally due regardless of future events. Those answers matter more than choosing a preferred label.

Evidence of alternative routes or contracts without the suspected dependency may reduce estimated exposure. Discovering additional links to the same operator may increase it. Record documented facts, assumptions and unresolved unknowns separately. A favorable opinion from one department does not close a material evidence gap. The assessment should be capable of changing when relevant evidence changes.

Negotiations with the assignor may involve additional documents, a revised pool, different collection procedures or reduced volume. These measures have different effects from a price adjustment. Higher compensation cannot resolve a prior assignment. Where the issue is temporary handling of receipts, a better-controlled payment route may address the problem more directly than a broad repurchase promise.

The final decision must fit the fund's policy, criteria and approval authorities. Record exposure before and after the decision, its rationale, outstanding conditions and responsibility for monitoring them. This lets investors distinguish a justified exception from relaxation driven by commercial pressure. Discipline includes limiting or rejecting a transaction even when the underlying entrepreneurial project appears economically attractive.

Common analytical mistakes and how to correct them

One recurring mistake is counting instruments as independent risks. Splitting an obligation into many installments adds rows without necessarily changing the payer or payment source. Consolidate exposure by debtor, economic group and relevant dependencies while preserving the maturity schedule. Apply the same care to legally separate businesses that share cash, control or a decisive buyer.

Another mistake is adding protections without accounting for overlap. If collateral support, co-obligations and repurchase promises rely on the same financial capacity, assuming full recovery from each exaggerates their combined benefit. Use a joint scenario: how much money would that source have, which claims compete for it and how long would recovery take? Include costs, disputes and payment priority rather than assuming instant enforcement.

Data errors can also change decisions. A claim may be assigned the seller's industry when a report intends to measure the debtor's; renegotiation may erase original arrears; reconciliation may count a receipt twice. Reconstruct the information, assess effects on limits and reporting, and determine whether earlier decisions were affected. Corrections need a traceable record, including when they reduce a previously disclosed indicator.

Finally, contracted income is not the same as cash received. The example anticipated R$360,000 but received only R$256,000 in the adverse scenario. Announced recoveries also differ from paid recoveries. Seek reconciliation among forecasts, accounting recognition and cash for each statement's purpose. A material error should prompt appropriate communication and review of the control that allowed it.

Finish with questions that can change the decision

Begin the conclusion by defining the assets: which receivables are acquired, how existence and ownership are established, and which conditions remain to be met. Then check consistency with fund policy, acquisition criteria and relevant limits. This sequence prevents a diversification narrative from being built around assets that could not properly belong in the portfolio under examination.

Organize the remaining investigation around economic dependencies, legal availability and operational conversion into cash. For each, identify the largest common loss source, available evidence and planned response. A portfolio with well-distributed buyers may need better assignment controls; another with strong documentation may have excessive maturity concentration. The needed correction depends on the actual weakness.

Require an understandable calculation stating exposed principal, income forgone, timing, recovery and cost without duplicating losses. In the exercise, R$2.6 million of common exposure produced R$1.04 million of principal loss and ending cash of R$9.116 million after income and expenses. Those results follow from stated assumptions. Changing recovery, timing or contagion changes the conclusion and may change the decision.

CVM Resolution 175 and Annex II are reference entry points alongside structure documents and appropriate professional analysis. Public securities distribution and regulated financial activities require appropriate structures and providers; technology or a CNAE business activity code is not permission. These references do not establish KEZS authorization. An informed decision recognizes costs, conflicts, uncertainty, possible loss and liquidity constraints, with no guarantee of approval, repayment, profitability, liquidity or capital protection.

Frequently asked questions.

References and context

Institutional reference points for checking applicable rules and guidance. Calculations and scenarios are original examples, not quotes, historical results or individualized recommendations.

• Banco Central — correspondentes no País ↗
• CVM — Resolução 88, texto e alterações ↗
• CVM — Resolução 175, fundos de investimento ↗
• Portal do Empreendedor — perguntas sobre MEI ↗
• ANPD — orientações e perguntas frequentes ↗

AI-assisted educational content. Decisions require verification of the information, instruments and providers in each actual case.